The Governance of the International Monetary Fund

THE GOVERNANCE OF THE INTERNATIONAL MONETARY FUND

ARIEL BUIRA

At the end of World War II the Bretton Woods conference gave birth to the International Monetary Fund (IMF) and the International Bank for Reconstruction and Development—better known as the World Bank. These two international financial institutions have come to exert a major—some would say dominant—influence on economic policy in developing countries over the last half century. During the nineties these two institutions have placed significant importance on governance issues among their member countries. The IMF, in particular, has given increased attention to such issues, following the approval of the Guidance Note on governance by the Executive Board1 five years ago (IMF 1997). The promotion of transparency and accountability are now at the core of the IMF’s efforts to ensure the good use of public resources as well as the domestic ownership of IMF programs (IMF 2001b). In this respect, transparency and accountability are important global public goods in an increasingly interdependent and democratic world. These two factors facilitate trust and confidence, bol- stering cooperation within the context of the international financial system. In recent years the IMF has developed and applied its instruments for pro- moting these objectives to an extent well beyond what was envisaged at the time the Guidance Note was approved. Indeed, the IMF helps countries identify any weaknesses that may exist in their institutional and regulatory frameworks that could give rise to poor governance; it then provides support in the design and implementation of remedial reforms. Given the strength of vested interests that benefit from the lack of transparency and accountability in these situations, over- coming these weaknesses often requires that the countries undertake significant structural reforms. By the very nature of its work then, the IMF exerts considerable influence over the majority of its 183 member countries, on such economic

Preparation of this chapter was financed in part by a grant from the OPEC Fund for International Development. It is a condensed version of a paper presented at the September 2000 meeting of the Intergovernmental Group of 24 on International Monetary Affairs and is based on and develops several themes in a paper prepared for the United Nations Development Programme (UNDP).

225 226 POLITICS and politically sensitive matters as wage policies, taxation and public expenditure levels, public sector prices and tariffs, subsidies and pensions, privatisation policies, the exchange regime and the exchange rate, interest rates and monetary policy, trade policy, financial sector regulations and others. With resources of over $280 billion and an expanded mandate, the IMF is probably today the most powerful of all international institutions. In view of its influence, it is of interest to consider to what extent the IMF’s own governance meets the standards of transparency and accountability required to ensure the ownership of programs by member countries and the prudent and effective use of international public resources. This chapter takes up that question, and begins with an analysis of the IMF’s power structure and of the issues raised by its current patterns of decisionmaking. Given the similarities between the IMF’s and the World Bank’s composition of shareholding and the resulting decisionmaking structures, this analysis of the IMF can also be applied to the World Bank.

CURRENT PATTERNS OF DECISIONMAKING

The decisionmaking patterns of the IMF member countries involve three key areas: the distribution of voting powers, the rules for decisionmaking, and the management structure within the IMF. Each of these dimensions is examined below in terms of its potential and actual effects on IMF policies and operations.

Voting powers Two aspects are remarkable in the distribution of voting power in the IMF. One is the skewed distribution of voting rights between industrial and developing countries, due in part to the diminished role of basic votes relative to quota-based votes. The other is that some of the variables used to determine quotas—a cru- cial element of voting powers—have not changed in more than 50 years. Both facts suggest that voting powers have not kept pace with changes in the global economy, undermining the IMF’s capacity to pursue its cooperative mandate.

Basic votes and quota-based votes. IMF members do not have equal voting power. Instead, they have weighted voting, a departure from the traditional practice of international organizations. To clarify, the vote of an IMF member has two com- ponents. Each member has 250 basic votes simply by virtue of its membership, as a symbolic recognition of the principle of the legal equality of states. Each member also has one additional vote for every 100,000 Special Drawing Rights (SDRs) of its quota. Because the number of basic votes has not changed with successive quota increases, the ratio of basic votes to total votes has declined from 12.4 percent of the voting power of the countries participating in the Bretton Woods con- THE GOVERNANCE OF THE INTERNATIONAL MONETARY FUND 227 ference (IMF 1993, schedule A) to 2.1 percent today, despite the entry of 135 new member countries. In fact, as a proportion of the total, the basic votes of the original members have declined from more than 12 percent to less than 0.4 percent as a result of a 37-fold increase in total quotas. This has changed the power structure of the IMF since the importance of the basic vote of a country is inversely related to the size of its economy, as basic votes represent a substantially higher proportion of the voting power of small countries. To illustrate: A country with a quota of 10 million SDRs would be entitled to 350 votes—100 votes due to its quota size and 250 basic votes for being a member. If the size of quotas is multiplied by ten, the country will have 1000 votes on account of its quota and 250 basic votes, for a total of 1250 votes. Thus the share of basic votes declines from over 70 percent to 20 percent of the total. Recall that in 1945 there were 14 countries—almost a third of the membership—whose quota was $10 million or less, and 28 countries—more than half the total—whose quotas were $50 million or less. With the passage of time, inflation and economic growth have combined to increase the size of the quotas. But since the number of basic votes has remained constant, their relative proportion to the total has declined, emasculating the role of basic votes and the relative influence of developing countries.

Determination of quotas. Because members’ quotas are the main factor determining voting rights, the process for setting such quotas should also be examined. It has been said that the quotas of the United States, the United Kingdom, the Soviet Union, and China were politically determined at the Bretton Woods conference. Raymond Mikesell, who was asked by the U.S. Treasury to estimate the first quotas, writes:

In mid-April 1943, White [i.e. Harry Dexter White, chief international economist at the U.S. Treasury in 1942–44] called me to his office and asked that I prepare a formula for the . . . quotas that would be based on the members’ gold and dollar holdings, national incomes, and foreign trade. He gave no instructions on the weights to be used, but I was to give the United States a quota of approximately $2.9 billion; the United Kingdom (including its colonies), about half the U.S. quota; the Soviet Union, an amount just under that of the United Kingdom; and China, somewhat less. He also wanted the total of the quotas to be about $10 billion. White’s major concern was that our military allies (President Roosevelt’s Big Four) should have the largest quotas, with a ranking on which the president and the sec- retary of state had agreed....I confess to having exercised a certain amount of freedom in making these estimates in order to achieve the predetermined quotas. (1994, pp. 22–23) 228 POLITICS

Subsequently,at the meeting of the Committee on Quotas,Mikesell was asked to explain the basis for his quota estimates, and he further writes:

I had anticipated this request and gave a rambling twenty-minute seminar on the factors taken into account in calculating the quotas, but I did not reveal the formula. I tried to make the process appear as scientific as possible, but the delegates were intelligent enough to know that the process was more political than scientific. (1994, pp. 35–36)

Given these historical facts, it is remarkable that—with only some adjustments in the weighting and definition of the main variables—the IMF contin- ues to use the original formula for determining members’ quotas. It is certainly understandable that the lack of equity and rationality in the quota criteria con- tinues to cause controversy and mistrust among members today, just as it did 50 years ago. The original formula is now combined with four other formulas, which give different weights to the same variables, and an element of discretion is used in selecting the formula to be applied in each case. (At times, the average of the various calculations is used to set a country’s quota.) It is therefore not surprising that current quotas are far from representative of the actual sizes of economies—of their ability to contribute to the IMF or of their importance in the world economy. This can be easily illustrated by the fact that such large countries as Brazil, Mexico, and the Republic of Korea, with real GNPs and populations much larger than those of Belgium, the Netherlands, and Switzerland, had quotas in 1999 that were only a fraction of those countries’ and fewer votes (table 1). Thus their share in decisionmaking is not commensurate with the systemic importance of their economies. It would be difficult to argue that the quota of China, the world’s second largest economy in purchasing power parity (PPP) terms, should be smaller than that of the Netherlands and similar to that of Belgium. Or that Belgium’s quota should be 52 percent larger than that of Brazil and 78 percent larger than that of Mexico. Moreover, it appears that many of the major differences arise between the quotas of industrial and developing countries and are not simply the result of history. For Switzerland, which recently joined the IMF, the quota was determined in line with those of industrial countries with similar economic structures and levels of development. As a result the distribution of quotas is skewed, as more recent quota numbers show (table 2). Quotas are important not just because they confer decisionmaking power but also because they determine access to financing. But for some exceptional cases, a member can borrow only up to a total of 300 per cent of its quota under regu- lar facilities. Thus the small quotas of developing countries limit both their share of voting power and their access to IMF resources.The consequences of the imbal- THE GOVERNANCE OF THE INTERNATIONAL MONETARY FUND 229

TABLE 1 IMF quotas and GNPs for selected countries Quota, effective January 1999a Purchasing power GNP, 1998 (billions of Special parity GNP, 1998 (billions of Country Drawing Rights) (billions of U.S. dollars) U.S. dollars)

Russian Federation 5.945 580.3 337.9 Netherlands 5.162 339.3 388.7 China 4.687 3,983.6 928.9 Belgium 4.607 239.7 259 Switzerland 3.458 189.1 284.8 Brazil 3.036 1,021.4 758 Mexico 2.586 785.8 380.9 Denmark 1.643 126.4 176.4 Korea, Republic of 1.634 569.3 369.9 a. Following the IMF’s Eleventh General Review of Quotas. Source: IMF, various issues, International Financial Statistics; World Bank, various issues, World Development Report.

TABLE 2 Distribution of IMF quotas by country group, 2001 Special Drawing Rights Share of total Country group (millions) (percent)

24 industrial countries 130,567 61.4 Oil-exporting countries 20,307 9.6 Non-oil-exporting developing countries 61,527 29.0 Total 212,401 100.0

Source: IMF, various issues, International Financial Statistics. ance of power have been further aggravated by the fact that since the late 1970s no industrial country has resorted to IMF support. This has changed the nature of the IMF: it has gone from being a credit cooperative from which all members draw resources from time to time, and therefore have an interest in credit being available on reasonable terms and conditions, to being an institution formed by two distinct groups of countries—industrial country creditors and developing country debtors. Hence, the fact that for over twenty years the IMF has only lent to developing countries has come to mean that the creditor countries try to lend as little as possible and therefore favor a hardening of conditionality, while the 230 POLITICS borrower countries, generally wanting to have ample access to financing on easy terms, tend to defend their short-term interests. The objectivity and impartiality of the Board,2 assumed by the Articles of Agreement has been eroded to a significant extent.3 More troubling, some past changes in the quotas of the main industrial countries were not based simply on the formulas—questionable as they may be—but on political criteria. In the Ninth General Review of Quotas, for example, Germany and Japan were assigned the same quota (giving both countries the second largest quotas after the United States) even though at the time Japan’s economy was twice the size of Germany’s. Similarly, France and the United Kingdom were given the third largest quotas even though Italy’s economy was larger than the United Kingdom’s. More recently, the IMF approved China’s request to increase its quota (following the resumption of Chinese sovereignty over Hong Kong), which has now been made equal—to the last decimal point—to Canada’s: 6,369.2 million SDRs. Yet one would be hard pressed to find similarities between their two economies; China’s economy is larger than Canada’s whether measured in purchasing power parity terms or in terms of market exchange rates (see table A.2 in the appendix). The IMF’s imbalanced allocations of quotas and voting powers necessarily raise questions about the legitimacy of access to its resources and of the decisions it makes.

Decisionmaking rules The IMF’s Articles of Agreement stipulate that some decisions require a qualified majority of the votes cast; that is, a particular proportion of the votes. At the Bretton Woods conference it was proposed that qualified majorities should be required in only two cases (one being quota adjustments), yet the subsequently accepted Articles of Agreement required qualified majorities for decisions in nine areas.With the first amendment to the Articles of Agreement, the number of these decisions rose to 18; with the second amendment, to 53. Forty of these are Executive Board decisions; 13 are Board of Governors4 decisions. The obvious explanation for this increase is the desire to protect some particular interest that might be affected by such decisions; decisions subject to a qualified majority can be taken only with the consent of the members having a high proportion of the total votes. Currently, the United States has 17.16 percent of total votes, Japan 6.16 percent, Germany 6.02 percent, and France and the United Kingdom 4.97 percent each, for a combined total of about 39 percent.5 If the votes cast by the Belgian, Canadian, Dutch, Finnish, Italian, and Swiss executive directors are added, the total exceeds 60 percent (IMF 2001a). The result is that decisions on 18 subjects requiring 85 percent of the total vote can be vetoed by one member country alone. Decisions on 21 other questions that require a 70 percent majority can be collectively vetoed by the five countries with the most voting power. THE GOVERNANCE OF THE INTERNATIONAL MONETARY FUND 231

Special majorities have been used to block decisions supported by an absolute majority of votes on increases in the size of the IMF (that is, quota increases) and in SDR allocations, sales of the IMF’s vast gold holdings, and policies on access to IMF resources. This special-majority requirement often has the effect of inhibit- ing even the discussion of important issues that would be difficult to resolve. The developing countries have argued that because voting itself is weighted—a situation that favors the industrial countries in decisionmaking—there should be no need for special majorities. However, the countries that for various reasons have favored such majorities have not been prepared to do away with them.

Management structure In principle, the staff and management of the IMF are subject to the political control of—and accountable to—the Board of Governors and its representatives, the executive directors. Thus the line of control and supervision runs from the Board of Governors, formed by the ministers of finance and central bank governors of member countries, to the executive directors who represent them, to the IMF management whom the executive directors appoint, and finally to the staff whom the management supervises. This management structure has three main weaknesses, involving the selection of the managing director, the role and functions of the executive directors, and the composition and background of the IMF staff.

The managing director. Formally, the Executive Board, on which all member countries are represented,appoints the managing director.However,there is an unwrit- ten understanding among major industrial countries by which the United States appoints the president of the World Bank, while Europe appoints the managing director of the IMF.As a result a handful of European officials—primarily British, French, and German—feel that it is their prerogative to appoint the managing director, with the consent of the United States but little consultation with the rest of the membership.The widely publicized discussions and disagreements between the U.S. and German governments leading up to the appointment of the managing director in 2000—with a touch of black comedy, as the United States rejected the first German candidate—should have ended all illusions about the participation of most countries in the process.But as the New York Times editorialized,“The managing director is too important to be chosen in secret by a few self-selected European countries” (22 November 1999). The selection process has to be opened up, considering the pivotal role the managing director plays in leading the IMF. Candidates for the position should state what their policies would be and how they would guide the IMF to attain its purposes. Since IMF operations are entirely with developing countries and transition economies, it is neocolonial to assume that only a European is capable of becoming managing director. It is also entirely implausible to suppose that there is no highly qualified developing country national who could take the position. 232 POLITICS

The issue is doubly important because the managing director and senior staff can in practice be held accountable only by the governments of a handful of countries.

The executive directors. Executive directors play two roles: they are national representatives, and they are IMF officials whose salaries are paid by the IMF. On the one hand they collectively determine the policies under which the organization is run and appoint the managing director, who in turn appoints and supervises the staff. On the other hand they represent member countries. However, directors representing developing countries—most of which turn to the IMF for financial support from time to time—have a limited ability to hold the staff and, much less, the management accountable, for two reasons. The first reason for this is their limited voting power, which means that they have little say on staff promotion or removal. No less important, the second reason is that being the representatives of petitioner governments limits these directors’ ability to question the staff. They are particularly limited in their ability to question staff in their area departments, since they have to rely on these same staff to prepare the papers presenting their (developing country directors’) countries’ case for financial support to the Board. Not wishing to diminish their own effectiveness in securing financial support for the countries they represent often means directors will not challenge or antag- onize senior staff, much less management, on whose judgment and goodwill their countries have to rely. A sensitive issue from the standpoint of transparency and accountability arises with the requirement of prior action or “pre-conditions” imposed by the staff and management on a country requesting a program, without the knowledge of the Board. In practice, the ability of developing country directors to exercise effective control over staff and management is seriously impaired. Indeed, directors who try to exercise their supervisory role run the risk that the staff or management complain about them to their authorities at the time financial support is negotiated, giving rise to a particularly delicate situation for directors from third countries,since the confidence of the authorities in them may be undermined.6 Furthermore, given their limited voting power, developing countries are forced to join other countries to muster a sufficient number of votes to elect an executive director to represent them on the Board. Consequently, developing country constituencies or “chairs” representing several countries usually rotate the positions of executive director and alternate executive director every two years among the several member countries they represent. While this practice permits a wide access of countries to the Board, it has two serious disadvantages. The first is that the rotation often means that newly appointed directors are not immedi- ately familiar with the complexities and the policies of the IMF; they often need a year to become familiar with the modus operandi of the institution. It follows THE GOVERNANCE OF THE INTERNATIONAL MONETARY FUND 233 that in policy discussions they are at a disadvantage relative to the staff and to industrial country directors, with their greater tenure and experience. The second disadvantage of this rotating arrangement is that the staff, whose appointments are permanent, know that these directors will normally depart at the end of the two-year cycle. So if the directors request changes in the presentation of annual consultation reports (or on policy matters) the staff do not favor, the staff can simply wait them out until the Director in question departs from the scene. Finally, in line with the imbalanced distribution of quotas noted earlier, country representation on the Board is necessarily skewed. Thus, while 24 industrial countries, none of which has an IMF-supported program, are at any one time represented by 10 or 11 executive directors—who generally receive considerable technical support from specialized offices in their capitals and are able to devote much of their time to policy issues—42 African countries (excluding Arab countries) are represented by only 2 executive directors. Consequently, those Directors each representing 20 countries or more are barely able to attend to the copious amount of bilateral business with the IMF of the countries they represent, several of which may be engaged in programs or the negotiation of programs at the same time. This provides little time to devote to the consideration of systemic policy issues.7

The staff. The IMF staff includes nationals from most (127 of the 183) member countries, but there has been a long-standing predominance of industrial country nationals among management and senior officers. These accounted for 26 of 31 such officials in 1996, improving slightly to 22 of 29 in March 2001 (IMF, various years). Moreover, since numerous developing country nationals in senior positions went directly from a US university to a position at the IMF, they cannot be said to have brought the experience and sensibility that come from work in their own countries. In addition, training in economics at the better graduate schools in Canada, the United Kingdom, and the United States—common among a large proportion of the staff—provides remarkable homogeneity in economic thought. This common approach facilitates IMF operations, but at times it may lead to a certain lack of pragmatism and innovation that creates difficulties in different environments.8

POLICY IMPLICATIONS

Clearly, the current power structure places a small number of countries in a dominant position, impairing the objectivity of IMF decisions and recommendations. In light of the power structure of the IMF and keeping in mind that economic policy is not an exact science, it is inconceivable that the staff are not influenced by the interests of the major quota-holding countries. 234 POLITICS

For example, the dramatic volatility of capital flows and the high costs of the Mexican crisis of 1994–95 should have been sufficient to lead the IMF to take a careful second look at the risks implied by full capital account liberalization and the integration of developing countries (particularly those without strong bank- ing systems) into international capital markets. The IMF continued to vigorously pursue the amendment to the Articles of Agreement to demand the opening of the capital account of developing countries. Only well after the Asian crisis of 1997–98 did the IMF concede that such crises do raise questions about the desir- ability of completely free capital movements and full capital account convertibil- ity (IMF 2000a). This policy shift came only after wide recognition of and broad public discontent with serious mistakes in the IMF’s performance during the Asian crisis (Sachs and Radelet 1998; Bhagwati 2000). However, belated correction of ineffective policies is not the only symptom of this problem. Technically questionable programs have sometimes been approved in order to support governments allied with the interests of the dominant country or countries, thereby placing the resources of the international community at risk. Furthermore,these cases have a demoralizing effect on the IMF staff,who are made to recognize that there are “special cases” based on non-economic considerations (Bordo and James 2000; Krueger 1997). The effectiveness of the staff could therefore be compromised, as they may impose a degree of self-censorship.9 Thus, it is difficult, if not impossible, to examine and analyze objectively initiatives or proposals that go against the interests of the major industrial countries. These facts cause even more concern given the international financial environment. In recent years, with the increase in capital mobility, developing countries have become much more vulnerable than in the past. They have frequently faced massive capital outflows leading to financial crises. The financial support required in such cases is much larger than that necessitated by traditional bal- ance of payments crises and greatly in excess, both in absolute terms and as a proportion of quotas, of that contemplated by IMF policies. Although in a number of well known cases such exceptional support has been forthcoming— Mexico received $48 billion of which SDR 12.3 billion ($17 billion) came from the IMF; the Republic of Korea $57 billion, of which SDR 15.5 billion ($22 billion) came from the IMF; Indonesia $43 billion; and Thailand $17 billion—this support has been decided on an ad hoc, discretionary basis by major industrial countries outside the framework of IMF policies. Of course, such arrangements are unlikely to comply with the principle of equality of treatment for all members. Moreover, it would seem that countries with the largest quotas—the creditor countries of the IMF—have opted to reduce their relative contributions and their exposure to IMF borrowing, thereby reducing the size of the IMF relative to world trade (table 3).As a result IMF quotas were equivalent to barely 6 percent of world imports in 1998, compared with almost 60 percent in 1944. Consequently, the THE GOVERNANCE OF THE INTERNATIONAL MONETARY FUND 235

TABLE 3 IMF quotas as a proportion of world imports and GDP,selected years, 1944–98 (percent) 1944 1950 1965 1970 1978 1990 1998

Quotas as a proportion of imports 58 17 15 14 9 6 6 Quotas as a proportion of GDP 4 2 2 2 1 1 1

Source: IMF 2000b, table 7.

IMF has inadequate resources to provide sufficient credit to member countries suffering payment imbalances. As IMF resources have not kept pace with financing needs, countries do not know how much, if any, financial support may be forthcoming In fact, at times of crisis it is very difficult for any country to obtain significant financing from the markets and normally countries cannot. In such circumstances, in the absence of sufficient financial support from the IMF,bilateral assistance may come with conditions and strings attached that have no bearing on the resolution of the crisis. For example, support for the Republic of Korea was made conditional on that country implementing more than ninety structural reform measures, including allowing foreign investors to purchase Korean businesses, opening the domestic financial sector to foreign banks and insurance companies, and liberalizing imports of Japanese cars. Regarding this issue, Feldstein (1998, p. 4 of electronic copy) writes,“A nation’s desperate need for short-term financial help does not give the IMF the moral right to substitute its technical judgments for the outcomes of the nation’s political process.” As has become clear, the governance of the IMF falls short of its standards and recommendations for transparency and accountability in the programs of its member countries. Transparency requires that decisions be the result of open discussions with broad participation. Accountability requires that those making decisions face up to their consequences. Legitimacy requires that the views and interests of all IMF members be given appropriate consideration. Only with transparency, accountability, and legitimacy can international institutions like the IMF hope to reconcile each country’s political and economic objectives with the international community’s wider interests, including the provision of global public goods like financial stability and market efficiency. These objectives will not be attained if decisions are made by a small group of industrial countries meeting outside the purview of the IMF or if the power structure in the IMF is decidedly imbalanced. This situation must be rectified. 236 POLITICS

OPTIONS FOR GOVERNANCE REFORM

The decisionmaking patterns in the IMF illustrate a problem facing other multi- lateral institutions: the IMF was founded during an era in which most developing countries were still under colonial rule and economies were less interdependent than today. Initially evolving under a marked east-west divide, the international political and economic environment has now achieved unprece- dented integration that has led more countries to seek membership in multilat- eral institutions such as the IMF. But the expansion of the IMF’s membership has not led to a broadening of its decisionmaking base. Recognizing these potential disparities, in 1999 the IMF’s managing director asked an independent group of external experts—led by Richard Cooper, professor of economics at Harvard University—to provide the Executive Board with a report on the adequacy of the quota formulas, including proposals for changes where appropriate. But the “Cooper report” (IMF 2000b) left unresolved many issues pertaining to IMF governance. Thus the following reforms—restructuring the Executive Board, revising quota formulas, and restoring the role of basic votes—are offered in hopes of contributing to a more effective IMF.

Restructuring the Executive Board The representation at the Executive Board could be undertaken in a way that an increase in the number of directors representing developing countries be matched by an equal reduction in the number of directors from industrial countries. The region with the greatest number of representatives on the Board is Europe, which currently holds eight chairs. Thus it is the obvious candidate for a reduction in the number of the chairs it holds. An additional reason for suggesting a reduction in the number of European directors is the process of unification that has resulted in a monetary union among 12 European countries,all now with a common interest rate and exchange rate policy. While one might think that all members of the European Monetary Union could be represented by one director, it would suffice to reduce the number of their directors to, say, two or three.10 Furthermore, in order to be able to give adequate attention to the needs of countries represented, no executive director should represent more than, say, 12–15 countries. In addition, the staff in the offices of executive directors representing more than one country should be strengthened significantly, in proportion to the number of countries represented. These measures should permit directors representing large constituencies to play a more active and effective role in policy discussions. While important, increased voice at the Board for developing countries is not by itself sufficient. This author recalls occasions when directors representing a major industrial country would not engage in the discussion of an issue that they could lose on grounds of logic. The directors would simply state that, after lis- THE GOVERNANCE OF THE INTERNATIONAL MONETARY FUND 237 tening to the arguments, they had not changed their position on the issue. Thus restructuring of the Executive Board should be accompanied by other reforms.

Revising quota formulas A more technical aspect of reform relates to quota formulas. The work of the Quota Formula Review Group (QFRG), which produced the Cooper report, has prompted consideration of the issues involved in the revision of the quota formulas and of the variables that should be included. Despite the shortcomings of the formula proposed, the initiative for the simplification and increasing the transparency of formula posited by the QFRG has considerable merit. However, the proposals from the QFRG can be improved upon. Some suggestions are presented below: • Relate total quotas to world trade and capital movements or to world GDP. A first approach would be to ensure that the size of the IMF should not fall below an agreed proportion of world trade or of world GDP. Note that simply establishing a ratio of say 15 percent of imports would more than double IMF resources, enabling it to reduce the costs of adjustment to members, making the institution far more relevant to their problems. Total quotas could be adjusted more or less automatically at three-year intervals to keep them from lagging significantly behind the expansion of the international economy. Additionally, the pattern of capital flows to prospective borrowing countries—all the IMF members except some 22 industrial countries—could also be considered in determining countries’ potential need for IMF support.11 • Include additional elements in the variable that measures the external vulnerability of countries. The inclusion in the quota formula of a measure of openness of the economy and of the dependence of countries on international financial markets would appear to be necessary, considering the volatility of short-term capital flows, which as is widely recognized, has been the determining factor in the financial crises suffered by emerging market economies over the last few years. • Use PPP-based GDP estimates in the quota formulas. For computation of the quotas, the figures used should be the PPP-based GDP estimates, since these are more stable and are unbiased compared to market exchange rate-based GDP. (See the appendix for a more thorough discussion of this issue.) More importantly, this should help correct the current underestimation of the economic size of developing countries and emerging market economies in the current quotas, as well as improve their representation at the Executive Board. There is no doubt that given the small size of quotas to GDP and that only a portion is contributed in hard currency (U.S. dollars), virtually all countries will be able to contribute to the IMF—which in any case is part of their reserves.12 Broadening the stake of developing countries in the IMF should also increase their contribution and lessen the concern of current creditor countries over the risk of IMF credits. 238 POLITICS

Restoring basic votes to their original function This reform measure could be initiated by increasing basic votes to an agreed proportion of total voting rights, say, 20 percent. Provision should then be made that in the future, basic votes will increase in the same proportion as total quotas, in order to preserve its role in the decisionmaking structure. It is important to note that the preservation of the share of basic votes in the total would not be an exceptional practice among international institutions. For instance, Article 33-1 in the Articles of Agreement of the Asian Development Bank provides that the relative importance of basic votes will remain constant over time as a proportion of the total vote. Similarly, the Articles of Agreement of the Inter-American Develop- ment Bank provide that no increase in the subscription of any member will become effective if it would reduce the voting power of certain countries or groups of countries below a given percentage of the total (IMF 2000b).

CONCLUSION

The world has changed considerably since the Bretton Woods conference of 1944. Developing countries now account for a much larger share of the world economy, with China, India, Brazil, Mexico, and Indonesia among the world’s 15 largest economies measured in real terms. The Soviet Union has disappeared. Trade has grown beyond expectations and vastly expanded international capital markets have taken a major unforeseen role. During the past 20 years or so IMF operations have been conducted exclusively with developing countries and, recently, also with transition economies. Moreover, in recent years the IMF has extended its conditionality to issues of governance. The divide among IMF members has widened during this period. On the one hand is a small group of creditor industrial countries with a majority vote; on the other is the large number of mostly prospective debtor developing countries with a minority vote and limited influence on policies. It is hardly coinci- dental that while the need for support of a significant group of developing countries has risen the size of the IMF has shrunk relative to world trade, and even more in relation to international capital movements. Over the last 20 years countries’access to IMF resources has become less predictable,conditionality has become gradually more restrictive—even for the compensatory financing facil- ity—and SDR allocations have been suspended since 1981. Consequently, decisions on major IMF-supported programs are taken outside the IMF by a very small group of countries, on a discretionary basis, without rules. This power distribution raises questions on the legitimacy, transparency and accountability of IMF governance. Clearly, the industrial countries no longer regard the IMF as being the center of the international monetary system but treat it instead as a specialized agency that assists the developing countries. Therefore, availability of sufficient support THE GOVERNANCE OF THE INTERNATIONAL MONETARY FUND 239 cannot be relied on. With this approach Keynes would disagree: “This [the IMF] is not a Red Cross philanthropic relief scheme, by which rich countries come to the rescue of the poor,” he declared, “it is a piece of highly necessary business mechanism which is at least as useful to the creditor as to the debtor” (as quoted in Chandavarkar 1984, Introduction). More broadly, the pressures of short-term self-interest and political expedi- ency appear to have blurred the Bretton Woods vision of international cooperation as a means to improve the workings of the world economy. The notion that national goals are often best attained through international cooperation tends to be forgotten. This situation is unsatisfactory. To improve the governance of the IMF and the World Bank in terms of participation, transparency, and accountability—and to enable them to meet the new challenges of the world economy— reform of the shareholding and decisionmaking structures must be undertaken. In thinking of this reform process one must note that a wider and more bal- anced participation of member countries is not contrary to the strict application of sound economic policies in the context of the IMF’s crisis prevention and resolution work nor in the World Bank’s project-related development work. In fact, member countries’ ownership of IMF and World Bank programs, projects, and policies—as well as stake holding in the global financial system—require it.

APPENDIX.THE ADVANTAGES OF PPP-BASED GDP

There are two key reasons to favor the use of PPP-based GDP as the basis for quota estimation, rather than market exchange rate-based GDP.PPP-based GDP is more stable, and it is unbiased when compared to market exchange rate-based GDP:

More stable PPP-based GDP is a far more stable basis for the quota formulas when compared to market exchange rate-based GDP.Consider the range of the exchange rate fluctuations and misalignments among major currencies. Simply recall that the

TABLE A.1 Ratio of Japan’s GDP to U.S. GDP converted using various methods, selected years, 1985–96 (percent) Conversion method 1985 1990 1993 1996

Annual average exchange rate 33 54 67 62 Five-year average exchange rate 39 59 65 61 PPP-based estimate 35 41 41 40

Source: OECD data [http://www.oecd.org]. 240 POLITICS

exchange rate between the dollar and the euro has gone (since the latter’s introduction in January 1999) from $1.16 per euro to $0.89 per euro (as of January 2002), a variation of over 20 percent in a period of three years.13 This alone would introduce substantial distortions in market exchange rate conversions of GDPs measured in these currencies and in others linked to these currencies. The problem is only somewhat reduced but does not disappear with the use of three-year averages as proposed in the Cooper report (IMF 2000b). In the example of Japan and the United States below, even the use of five-year averages was not sufficient to eliminate significant fluctuations in market rate-based GDP estimates. Contrast these changes with the stability displayed by PPP-based GDP estimates (table A.1).

Unbiased The use of market exchange rates substantially undervalues the GDPs of developing countries, as the prices and wages prevailing in the tradable goods sector

TABLE A.2 PPP-based and exchange rate-based GDPs of selected countries,2000 Share of world GDP World ranking (percent) Exchange PPP-based Exchange rate- PPP-based rate-based Country GDP based GDP GDP GDP

United States 21.5 31.5 1 1 China 11.2 3.4 2 6 Japan 7.4 14.9 3 2 India 5.4 1.5 4 12 Germany 4.6 6.0 5 3 France 3.2 4.1 6 5 United Kingdom 3.1 4.5 7 4 Italy 3.0 3.4 8 7 Brazil 2.9 1.9 9 9 Russian Federation 2.7 0.8 10 17 Mexico 2.0 1.8 11 10 Canada 1.9 2.2 12 8 Korea, Republic of 1.8 1.5 13 13 Spain 1.7 1.8 14 11 Indonesia 1.4 0.5 15 26 Netherlands 0.9 1.2 19 15

Source: World Bank, World Development Indicators Database, 2001. THE GOVERNANCE OF THE INTERNATIONAL MONETARY FUND 241 are higher than those prevailing in the non-tradable goods sector—a phenome- non that is not significant in industrial countries. This represents a major distor- tion inherent in the market exchange rate-based GDP,which would argue against its use in GDP comparisons between industrial and developing countries. Since PPP-based GDP estimates on the other hand, do not introduce a measurement bias against any group of countries, they would appear to be preferable for this purpose. To visualize the importance of the differences simply consider table A.2 which clearly shows how exchange rate-based GDP is a striking underestimation of the size of developing countries like China and India.

NOTES

1. The Executive Board is responsible for conducting the day-to-day business of the IMF. It is composed of 24 executive directors, who are appointed or elected by member countries or groups of countries. Some directors represent more than one country. The managing director serves as its chairman. Meeting several times a week, the Executive Board deals with a wide variety of policy, operational, and adminis- trative matters, including surveillance of members’ macroeconomic policies, provision of IMF financial assistance to member countries, and discussion of systemic issues in the global economy. (See http://www.imf.org/external/pubs/ft/survey/sup 2001/index.htm#2) 2. From hereon, unless otherwise stated, the Board refers to the Executive Board. 3. According to the Articles of Agreement, quotas are to be reviewed by the Board of Governors at intervals of no more than five years and if appropriate adjusted (IMF 1993, Art. III Sec. 2a). However, some 70 percent of all increases have been across the board, producing a substantial inertia in quota shares and failing to adequately reflect changes that have taken place in the world economy since 1944. 4. The Board of Governors of the IMF consists of one governor and one alternate for each member country. The governor, appointed by the member country, is usually the minister of finance or the central bank governor. The Board of Governors has delegated to the Executive Board all except certain reserved powers. It normally meets once a year. (See http://www.imf.org/external/pubs/ft/survey/sup2001/index.htm#2) 5. In the World Bank the same five countries control 38 percent of the total votes. If the votes cast by the Belgian, Canadian, Danish, Dutch, Italian, and Swiss directors are added, the total is around 60 percent. (See http://www.worldbank.org/about/ organization/voting/librd.htm.) 6. The situation is no different in the World Bank. As Fidler (Financial Times, 28 August 2001, editorial page) writes, Although no other member states come close to matching US power over the bank, all its influential owners—Britain, France, Germany and others—have borrowing governments whose interests they purport to sponsor, as well as key issues (such as the environment) that are viewed as important by their elec- torates. Otherwise, accountability and scrutiny from donor governments are 242 POLITICS

uneven at best and non-existent at worst. Meanwhile, the few powerful borrowing nations that could exert some influence are afraid to voice their concerns lest they lose access to bank finance. 7. In the World Bank the problem may be worse given the large number and variety of projects and the fact that Sub-Saharan African countries are also represented by only two directors. 8. Again, this characteristic in the IMF can also be observed in the World Bank. As the World Bank’s official history states, The Americans had a secure enough lead in the Bank throughout the half century to help it avoid the clutter of country quotas in its hiring, to recruit per- sonnel on merit from the developed and developing countries alike, and to build a work force that some saw as comparatively denationalized and homoge- neous….One unmistakable factor that contributed to this homogenizing and that grew stronger over time was economics. As this work demonstrates, economics would become the Bank’s hallmark scholarly discipline, and the econo- mists who heavily shaped Bank operations as well as its research were recruited from an array of countries. To a large degree, however, they were the product of the graduate economics departments of English-speaking, but specially American, universities. This fact, as it played into the Bank’s consulting, research, technical assistance and agenda setting, would enhance the US role in the institution beyond the apparatus of formal governance (Kapur, Lewis, and Webb 1997). 9. For example, David Finch, former director of the IMF’s Exchange and Trade Relations Department, resigned under political pressure to relax IMF conditionality for Egypt and Zaire (“IMF Silent on Resignations,” Financial Times, 21 March 1987). 10. The cumulative GDP of the 15 countries of the European Union is roughly equal to that of the United States; however, the European Union has seven directors (as of the end of April 2001), and its cumulative quota is about 70 percent larger than that of the United States. 11. This would not preclude any industrial country from turning to the IMF for support. 12. One argument for the use of GDP based on market exchange rates is that it is a better measure of a country’s ability to contribute to the IMF. However, the rela- tionship between actual contributions as determined by quotas and the ability to contribute as a proportion of GDP is far from being a binding restriction for three reasons. First, quotas are a very small proportion of GDP: only 1 percent at the time of the Eleventh Quota Review in 1998 and an even smaller proportion today. Second, since conversion of GDP at market rates produces significantly smaller GDP estimates than PPP-based conversion, the potential contributions by developing countries are such a small proportion of their GDP that the argument loses significance. Third, note that only 25 per cent of the member’s contributions or quota is paid in foreign currencies. Taken together, these facts weaken the “ability to contribute” argument—the main argument against the use of PPP-based GDP—to the point at which it becomes irrelevant. THE GOVERNANCE OF THE INTERNATIONAL MONETARY FUND 243

13. Data are from the Pacific Exchange Rate Service, as provided by Professor Werner Antweiler of the University of British Columbia (http://pacific.commerce. ubc.ca/xr/).

REFERENCES

ADB (Asian Development Bank). 1966. “Agreement Establishing the Asian Devel- opment Bank.” Manila. [http://www.adb.org/Documents/Reports/Charter/ charter.pdf]. Bhagwati, Jagdish N. 2000. “Lessons from the East Asian Experience: Opening Address.” In Eric S. Rosengren and John S. Jordan, eds., Building an Infrastructure for Financial Stability. Conference Series, no. 44. Boston: Federal Reserve Bank of Boston. Bordo, Michael, and Harold James. 2000. The International Monetary Fund: Its Present Role in Historical Perspective. NBER Working Paper 7724. Cambridge, Mass.: National Bureau of Economic Research. Chandavarkar, Anand G. 1984. “The International Monetary Fund, Its Financial Organization and Activities.” International Monetary Fund, Washington, D.C. Feldstein, Martin. 1998. “Refocusing the IMF.” Foreign Affairs 77 (2): 20–33. IMF (International Monetary Fund). 1979. Guidelines on Conditionality. Decision 6056 (79/38). 2 March. Washington, D.C. ———. 1993. “Articles of Agreement.” Washington, D.C. [http://www.imf.org/ external/pubs/ft/aa/]. ———. 1997. “IMF Adopts Guidelines Regarding Governance Issues.” Washington, D.C. [http://www.imf.org/external/np/sec/nb/1997/nb9715.htm#I2]. ———. 2000a. “Country Experiences with the Use and Liberalization of Capital Controls.” Occasional Paper. Washington, D.C. ———. 2000b. “Report to the IMF Executive Board of the Quota Formula Review Group.” Washington, D.C. ———. 2001a. “Annual Report of the Executive Board for the Financial Year Ended April 30, 2001.” Washington, D.C. [http://www.imf.org/external/pubs/ft/ar/ 2001/eng/index.htm]. ———. 2001b. “Review of the Fund’s Experience in Governance Issues.” Washington, D.C. [http://www.imf.org/external/np/gov/2001/eng/report.htm]. ———. Various issues. International Financial Statistics. Washington, D.C. Kapur, Devesh, John P. Lewis, and Richard Webb. 1997. The World Bank: Its First Half Century. Washington, D.C.: Brookings Institution Press. Krueger, Anne O. 1997. Whither the World Bank and the IMF? NBER Working Paper 6327. Cambridge, Mass.: National Bureau of Economic Research. 244 POLITICS

Mikesell, Raymond F. 1994. “The Bretton Woods Debates: A Memoir.” Essays in International Finance 192. Princeton University, International Finance Section, Princeton, N.J. Sachs, Jeffrey, and Steven Radelet. 1998. “The East Asian Financial Crisis: Diagnosis, Remedies, Prospects.” Brookings Papers on Economic Activity. Washington, D.C.: Brookings Institution. World Bank. 2001. World Development Indicators. Washington, D.C. ———. Various years. World Development Report. New York: Oxford University Press.